Command A vs Llama 4 Maverick
Benchmarks, API pricing and specs, head to head. Data updated 2026-06-10.
Command A
Cohere · Mar 2025
Cohere's enterprise RAG specialist — 111B dense, two-GPU deployable, with downloadable weights for private deployments.
Llama 4 Maverick
Meta · Apr 2025
Meta's natively multimodal 400B MoE — the largest openly downloadable US-made model, served cheaply across many hosts.
The verdict
These two models don't yet share verified results on the benchmarks we track, so judge them on specs, pricing and intended use. On price the gap is dramatic: Llama 4 Maverick works out roughly 11x cheaper per blended million tokens. Llama 4 Maverick also takes 1M of context versus 262K for Command A.
Specs & pricing
| Command A | Llama 4 Maverick | |
|---|---|---|
| modhub Index | — | — |
| Input price / 1M | $2.5 | $0.27 |
| Output price / 1M | $10 | $0.85 |
| Context window | 262K | 1M |
| Max output | 8K | — |
| Open weights | yes (CC-BY-NC 4.0) | yes (Llama 4 Community License) |
| Reasoning model | no | no |
| Multimodal input | text | text, image |
| Knowledge cutoff | Jun 2024 | Aug 2024 |
| Released | Mar 2025 | Apr 2025 |
| Example monthly cost* | $40.00 | $3.98 |
* 10M input + 1.5M output tokens per month at list prices, no caching. Green = better value on that row.
Frequently asked questions
- Which is better, Command A or Llama 4 Maverick?
- These two models don't yet share verified results on the benchmarks we track, so judge them on specs, pricing and intended use. On price the gap is dramatic: Llama 4 Maverick works out roughly 11x cheaper per blended million tokens. Llama 4 Maverick also takes 1M of context versus 262K for Command A.
- Which is cheaper, Command A or Llama 4 Maverick?
- Command A costs $2.5/$10 per million input/output tokens, while Llama 4 Maverick costs $0.27/$0.85. For a typical workload of 10M input and 1.5M output tokens per month, that's $40.00 versus $3.98.
- Which model is better for coding, Command A or Llama 4 Maverick?
- We don't yet track SWE-bench Verified results for both models; check their individual pages for coding-related scores.